Solution of Wave Equation by Linear Regression Artificial Neural Network
DOI:
https://doi.org/10.21123/bsj.2024.11520Keywords:
Artificial Neural Network, Linear Regression, LM Training, Partial Differential Equation, Wave EquationAbstract
Wave partial differential equations are one of the most important topics that model, for example, the wave motion of ground vibrations. Hence, finding approximate solutions to such equations with high accuracy and speed faster than difficult and complex analytical solutions has become possible through the use of artificial intelligence and machine learning methods. This research has three goals. The first is to transform the problem of the initial value of the wave equation into its legal form and find its accurate solution. The second is to write a linear-regression neural network algorithm. The third result is applying this algorithm to find a numerical solution to the initial value problem under the study. The last step is to compare the solution using a table and figures for certain parameters and initial conditions values to demonstrate the efficiency of the artificial neural network method. Approximate solutions were obtained with an error amount close to zero compared to the real solution of the wave differential equation by applying the artificial neural network that represents the linear regression equation. This gives the advantage of high speed in obtaining the solution to this type of differential equation.
Received 08/05/2024
Revised 21/09/2024
Accepted 23/09/2024
Published Online First 20/12/2024
References
Nazarzadeh A, Eslami M, Mirzazadeh M. Exact solutions of some nonlinear partial differential equations using functional variable method. Pramana - J Phys. 2013; 81(2): 225–36. http://dx.doi.org/10.1007/s12043-013-0565-9.
Agrawal A, Gans JS, Goldfarb A. Exploring the impact of artificial Intelligence: Prediction versus judgment. Inf Econ Pol. 2019; 47:1–6. http://dx.doi.org/10.1016/j.infoecopol.2019.05.001
Li Z, Kovachki N, Azizzadenesheli K, Liu B, Bhattacharya K, Stuart A, et al. Fourier neural operator for parametric partial differential equations. arXiv [cs.LG]. 2021; 1-16. https://doi.org/10.48550/arXiv.2010.08895
Al-Saif NSM. Neural network for solving integro partial differential equation. In: 2022 2nd International Conference on Computing and Machine Intelligence (ICMI) 15-16 July 2022, Istanbul, Turkey. IEEE. 2022; 1-4. https://doi.org/10.1109/ICMI55296.2022.9873771
Grundland AM. On k-wave solutions of quasilinear systems of partial differential equations [Internet]. arXiv [math.AP]. 2024; (Special Issue 1): 1-20. https://doi.org/10.48550/arXiv.2305.04090
Adwan MI, Al-Jawary MA, Tibaut J, Ravnik J. Analytic and numerical solutions for linear and nonlinear multidimensional wave equations. Arab J Basic Appl Sci. 2020; 27(1): 166–82. http://dx.doi.org/10.1080/25765299.2020.1751439
Guo Y, Cao X, Peng K. Solving nonlinear soliton equations using improved physics-informed neural networks with adaptive mechanisms. Commun Theor Phys. 2023; 75(9): 095003. http://dx.doi.org/10.1088/1572-9494/accb8d
Hussain A, Uddin M, Haq S, Jan HU. Numerical solution of heat equation in polar cylindrical coordinates by the meshless method of lines. J Math. 2021; 2021: 1–11. http://dx.doi.org/10.1155/2021/8862139
Abubakar SS, Ibrahim M, Abubakar N. A study on some application of partial differential equation s in the heat transfer. Journal of Applied Physical Science International. 2019; 11(3): 88–94.
Khan NA, Khalaf OI, Romero CAT, Sulaiman M, Bakar MA. Application of Euler neural networks with soft computing paradigm to solve nonlinear problems arising in heat transfer. Entropy (Basel). 2021; 23(8): 1053-1057 . http://dx.doi.org/10.3390/e23081053
Pratama DA, Bakar MA, Ismail NB, Mashuri M. ANN-based methods for solving partial differential equations: a survey. Arab J Basic Appl Sci. 2022; 29(1): 233–48. http://dx.doi.org/10.1080/25765299.2022.2104224
Blechschmidt J, Ernst OG. Three ways to solve partial differential equations with neural networks -- A review. arXiv:2102.11802 [math.NA]. 2021; 1-32. http://dx.doi.org/10.48550/ARXIV.2102.11802
Hu Y, Zhao T, Xu Z, Lin L. Neural Time-Dependent Partial Differential Equation. Communications in Information and Systems. 2021; 1-25.
Hakan K.F Solving Partial Differential Equations with Neural Networks. Department of Mathematics, Uppsala University; 2023.
Poonia S. Solution of l differential equations using by Sumudu transform. Int J Math and computer research. 2013; 1(2): 316–323.
Mjthap HZ, Al-Azzawi SN. Some properties of Sumudu transformation. J Interdiscip Math. 2019; 22(8): 1543–7. http://dx.doi.org/10.1080/09720502.2019.1706853.
Amin R, Shah K, Asif M, Khan I. A computational algorithm for the numerical solution of fractional order delay differential equations. Appl Math Comput. 2021; 402(125863): 125863. https://doi.org/10.1016/j.amc.2020.125863 .
Amin R, Ahmad H, Shah K, Bilal Hafeez M, Sumelka W. Theoretical and computational analysis of nonlinear fractional integro-differential equations via collocation method. Chaos Solitons Fractals. 2021; 151(111252): 111252. https://doi.org/10.1016/j.chaos.2021.111252 .
Basheer IA, Hajmeer M. Artificial Neural Networks: Fundamentals, Computing, Design, and Application. Journal of Microbiological Methods. 2000; 43(1): 3-31. http://dx.doi.org/10.1016/S0167-7012(00)00201-3 .
Nguyen T-A, Ly H-B, Mai H-VT, Tran VQ. On the training algorithms for artificial neural networks in predicting the shear strength of deep beams. Complexity. 2021; 2021: 1–18. http://dx.doi.org/10.1155/2021/5548988.
Downloads
Issue
Section
License
Copyright (c) 2024 Nahdh S. M. Al-Saif Al-Saif , Mahmood A. Shamran , Samah Mohammed Ali Ali , Saad Naji Al-Azzawi
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite
